Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
FACIAL ACTION RECOGNITION MODEL TRAINING METHOD, FACIAL ACTION RECOGNITION METHOD AND APPARATUS, COMPUTER DEVICE, AND STORAGE MEDIUM
Document Type and Number:
WIPO Patent Application WO/2021/068325
Kind Code:
A1
Abstract:
A facial action recognition model training method, comprising: acquiring a facial action recognition data set, the facial action image recognition data set comprising multiple facial action images; inputting the facial action images in the facial action recognition data set into a preset multi-task convolutional neural network, and using the multi-task convolutional neural network to perform facial detection on the facial action images to obtain multiple corresponding facial feature images; according to a preset rule, respectively adding black blocks to the facial feature images, and using the obtained images as a training image set; inputting the training image set into a preset neural network to be trained in order to train the neural network to be trained, and using the trained neural network to be trained as a facial action recognition model.

Inventors:
LUO LINYAO (CN)
XU GUOQIANG (CN)
QIU HAN (CN)
Application Number:
PCT/CN2019/117027
Publication Date:
April 15, 2021
Filing Date:
November 11, 2019
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
PING AN TECH SHENZHEN CO LTD (CN)
International Classes:
G06K9/00
Foreign References:
CN107967456A2018-04-27
CN109840477A2019-06-04
CN107016370A2017-08-04
US20030009078A12003-01-09
Attorney, Agent or Firm:
ADVANCE CHINA IP LAW OFFICE (CN)
Download PDF: